1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
|
// Copyright Vespa.ai. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
package com.yahoo.vespa.model.application.validation.change;
import com.yahoo.config.model.api.ConfigChangeAction;
import com.yahoo.config.model.api.OnnxModelCost;
import com.yahoo.config.model.deploy.DeployState;
import com.yahoo.vespa.model.Host;
import com.yahoo.vespa.model.VespaModel;
import com.yahoo.vespa.model.container.ApplicationContainerCluster;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.Optional;
import java.util.Set;
import java.util.logging.Logger;
import static java.util.logging.Level.FINE;
import static com.yahoo.config.model.api.OnnxModelCost.ModelInfo;
/**
* If Onnx models change in a way that requires restart of containers in
* a container cluster this validator will make sure that restartOnDeploy is set for
* configs for this cluster.
*
* @author hmusum
*/
public class RestartOnDeployForOnnxModelChangesValidator implements ChangeValidator {
private static final Logger log = Logger.getLogger(RestartOnDeployForOnnxModelChangesValidator.class.getName());
@Override
public List<ConfigChangeAction> validate(VespaModel currentModel, VespaModel nextModel, DeployState deployState) {
if ( ! deployState.featureFlags().restartOnDeployWhenOnnxModelChanges()) return List.of();
List<ConfigChangeAction> actions = new ArrayList<>();
// Compare onnx models used by each cluster and set restart on deploy for cluster if estimated cost,
// model hash or model options have changed
// TODO: Skip if container has enough memory to handle reload of onnx model (2 models in memory at the same time)
for (var cluster : nextModel.getContainerClusters().values()) {
var clusterInCurrentModel = currentModel.getContainerClusters().get(cluster.getName());
if (clusterInCurrentModel == null) continue;
var currentModels = clusterInCurrentModel.onnxModelCostCalculator().models();
var nextModels = cluster.onnxModelCostCalculator().models();
log.log(FINE, "Validating " + cluster + ", current models=" + currentModels + ", next models=" + nextModels);
actions.addAll(validateModelChanges(cluster, currentModels, nextModels));
actions.addAll(validateSetOfModels(cluster, currentModels, nextModels));
}
return actions;
}
private List<ConfigChangeAction> validateModelChanges(ApplicationContainerCluster cluster,
Map<String, ModelInfo> currentModels,
Map<String, ModelInfo> nextModels) {
List<ConfigChangeAction> actions = new ArrayList<>();
for (var nextModelInfo : nextModels.values()) {
if (! currentModels.containsKey(nextModelInfo.modelId())) continue;
log.log(FINE, "Checking if " + nextModelInfo + " has changed");
modelChanged(nextModelInfo, currentModels.get(nextModelInfo.modelId())).ifPresent(change -> {
String message = "Onnx model '%s' has changed (%s), need to restart services in %s"
.formatted(nextModelInfo.modelId(), change, cluster);
setRestartOnDeployAndAddRestartAction(actions, cluster, message);
});
}
return actions;
}
private List<ConfigChangeAction> validateSetOfModels(ApplicationContainerCluster cluster,
Map<String, ModelInfo> currentModels,
Map<String, ModelInfo> nextModels) {
List<ConfigChangeAction> actions = new ArrayList<>();
Set<String> currentModelIds = currentModels.keySet();
Set<String> nextModelIds = nextModels.keySet();
log.log(FINE, "Checking if model set has changed (%s) -> (%s)".formatted(currentModelIds, nextModelIds));
if (! currentModelIds.equals(nextModelIds)) {
String message = "Onnx model set has changed from %s to %s, need to restart services in %s"
.formatted(currentModelIds, nextModelIds, cluster);
setRestartOnDeployAndAddRestartAction(actions, cluster, message);
}
return actions;
}
private Optional<String> modelChanged(OnnxModelCost.ModelInfo a, OnnxModelCost.ModelInfo b) {
if (a.estimatedCost() != b.estimatedCost()) return Optional.of("estimated cost");
if (a.hash() != b.hash()) return Optional.of("model hash");
if (a.onnxModelOptions().isPresent() && b.onnxModelOptions().isEmpty()) return Optional.of("model option(s)");
if (a.onnxModelOptions().isEmpty() && b.onnxModelOptions().isPresent()) return Optional.of("model option(s)");
if (a.onnxModelOptions().isPresent() && ! a.onnxModelOptions().get().equals(b.onnxModelOptions().get()))
return Optional.of("model option(s)");
return Optional.empty();
}
private static void setRestartOnDeployAndAddRestartAction(List<ConfigChangeAction> actions, ApplicationContainerCluster cluster, String message) {
cluster.onnxModelCostCalculator().setRestartOnDeploy();
actions.add(new VespaRestartAction(cluster.id(), message));
}
private static boolean enoughMemoryToAvoidRestart(ApplicationContainerCluster cluster) {
// Node memory is known so convert available memory percentage to node memory percentage
double totalMemory = cluster.getContainers().get(0).getHostResource().realResources().memoryGb();
double availableMemory = Math.max(0, totalMemory - Host.memoryOverheadGb);
double costInGb = (double) cluster.onnxModelCostCalculator().aggregatedModelCostInBytes() / 1024 / 1024 / 1024;
return ( 2 * costInGb < availableMemory);
}
}
|
